File(s) under permanent embargo

Reason: Restricted by author. A copy can be supplied under Section 51(2) of the Australian Copyright Act 1968 by submitting a document delivery request through your library or by emailing

Theoretical and practical insights for anorexia nervosa and major depression: novel neurobiological targets for pharmacology and brain stimulation therapies

posted on 22.03.2017, 01:46 by Keating, Charlotte
Major Depression (MD) and Anorexia Nervosa (AN) often present co-morbidly and both share neurobiological abnormalities. MD presents up to 3 times as often in females than males and AN presents in up to 95% of females. In the illness phase, pathophysiological evidence indicates similar abnormalities in both clinical groups including; dysfunction in the serotonin system (5-hydroxytryptamine, 5-HT) (of which some abnormalities persist following recovery) and between 60-80% of patients in both groups present with significant hyperactivity of the hypothalamopituitary adrenal (HPA) axis. The first line treatment approach for MD involves modulation of the 5-HT system using selective serotonin reuptake inhibitors(SSRIs) but these strategies are not highly effective in patients with MD, whereby up to 30-40% of patients present with treatment resistance. For AN, treatment with SSRIs has been shown to be less effective relative to MD. A commonly reported consequence of SSRI treatment, is the normalization of indices of the HPA axis (ie., cortisol, adrenocorticotropic hormone, ACTH), which is consistent with recovery levels in both clinical groups and suggests that a reduction in HPA axis concentrations may be associated with response to treatment and recovery. The mechanisms contributing to a lack of MD patient’s response to antidepressants are unknown. Several lines of empirical evidence, however, support an association between upregulated activity of the HPA axis (hyperresponsiveness) and failed treatment response in MD, but the nature of dysregulation of the HPA axis (eg., the contribution of central, hypothalamic mechanisms) to treatment response has not been investigated directly, precluding our capacity to quantify definitively the localised (central) impact of medication of brain neurotransmitters and neuropeptides. For instance, previous research has not been able to measure the central influence of the neuroendocrine system on treatment response in patients with MD, where published studies have only been able to measure peripheral metabolites and have only done so with patients treated on heterogeneous pharmacological treatment programs (including direct comparison between patients and healthy controls where patients were receiving either pharmacological and nonpharmacological treatment strategies). Therefore, an aim of this thesis was to experimentally understand using simultaneous internal jugular venous sampling (a technique enabling blood plasma sampling from brain turnover) and arterial (peripheral) sampling, the impact of treatment on stress indices including plasma cortisol and oxytocin concentrations. We showed that patients’ HPA axes (ie., plasma cortisol concentrations) respond heterogeneously to SSRI treatment; some showing elevations (hyperresponsive)others a reduction (hyporesponsive) in activity and others still, no change from pre to post-treatment. Hyperresponsive patients demonstrated on average, a 3 fold deficit in baseline plasma oxytocin concentrations relative to no change patients. Oxytocin concentrations were not influenced by treatment in any patient sub-group. Furthermore we showed in patients (and the same pattern in a small group of healthy controls), no difference in brain and peripheral plasma oxoytocin concentrations, suggesting for the first time that peripheral concentrations may be a reliable marker of central function. AN is associated with the highest mortality of any psychiatric illness and there are few theoretical models that have been developed, which guide effective treatment strategies. Unlike MD, relatively little is understood about brain mechanisms driving AN, and there are few studies suggesting a strong link between brain function and behaviour that give direction to the field for future research. Therefore, an aim of this thesis was to develop a novel theory encompassing a neurocircuit and relevant systems linked to empirically demonstrated abnormalities, that account for key clinical symptoms; excessive exercise, self starvation and body dysmorphic selfperceptions in patients with AN. This set of aims was achieved. Based on empirical evidence that patients with AN experience abnormally, reward, I developed a novel extension to this theory recently characterized as, reward conflict (see chapter 5). Where the traditional notion of abnormal reward function in AN has focused on the anhedonic hypothesis, where patients engage in self-starvation and excessive exercise because they induce reward based on the physiological finding that HPA axis activation induces dopamine (DA) release, reward conflict is an extension to this. Reward conflict is considered to result from continued, pathological and unrelenting engagement in illness behaviours. It is proposed that patients unrelenting engagement in pathological behaviors of self-starvation and excessive exercise might otherwise be expected to be associated with a sense of punishment, however, contamination of the experience of reward (with punishment) is proposed to preclude patients perceiving the punishment associated with their pathological behaviors. The reward conflict (or contamination) theory is proposed to reinforce and maintain patients’ illness behaviors. The anterior cingulate cortex has been rationalized as a primary locus for reward contamination in a seminal publication (chapter 5). This neuranatomical locus involved in reward contamination has been extended to encompass a neurocircuit for symptoms of AN (chapter 6 and 7). From the development of this neurocircuit, a novel, multimodal treatment approach involving brain stimulation therapy early in treatment and the potential for novel drug combinations, based on an innovative interpretation of empirically demonstrated deficits in various systems, and their relationship to clinical symptoms, consistent with reward conflict has been developed (chapter 6). In addition to a seminal publication for the theory of reward conflict in anorexia nervosa (chapter 5), a larger review, encompassing a neurocircuit as well as specific functional associations for systems-related deficits, are currently under review (chapter 7). The results of this thesis confirm practically and propose theoretically, that the HPA axis (and its pathological up-regulation) presents an overlooked system contributing to the reinforcement and pathophysiology of psychiatric illnesses including AN and MD that tend to present more often in women than in men. A key focus for future research will be better understanding the contribution of the HPA axis among other hormone mediators to AN and MD to realise the potential for improved targeting of the symptoms of both AN and MD via novel, neuroendocrine system targets and brain stimulation therapies.


Principal supervisor

Jayashri Kulkarni

Year of Award


Department, School or Centre

Psychology and Psychiatry

Campus location



Doctor of Philosophy

Degree Type



Faculty of Medicine Nursing and Health Sciences